Great stories have always helped us understand complex truths about human behavior, and sometimes the simplest tales reveal the most profound insights about how incentives shape our choices. In this exclusive Stankevicius article, the author interleaves insights from Martine Murray’s “The Wanting Monster,” a children’s fable, with an exploration of how America might realign Silicon Valley incentives to serve American workers and communities first. The author explored previously in exclusive Stankevicius articles: Autonomous AI’s Spellbook and the Absent Necromancer, and Forging the Future: President Trump’s AI Vision and OpenAI’s Challenge to DeepSeek.
The challenge facing America’s technology sector today stems from a fundamental misalignment of incentives that has driven companies to outsource essential AI work overseas while promising domestic job creation and technological leadership. As economist Steven Landsburg observed, “People respond to incentives. The rest is commentary.”
President Trump’s AI Vision
“What had been withheld was released; what had dried up, flowed.” Martine Murray, The Wanting Monster
President Trump’s vision for American AI dominance represents a bold reimagining of technological leadership on domestic soil. Through his ambitious $500 billion Stargate project with OpenAI, Oracle, and SoftBank, the President’s vision promises to create “over 100,000 American jobs almost immediately” and establish the United States as the undisputed leader in artificial intelligence development. His Executive Order Removing Barriers to American Leadership in Artificial Intelligence eliminating Biden Administration AI policies aims to enhance “America’s global AI dominance” while ensuring “American development of AI systems must be free from ideological bias or engineered social agendas.”
Yet the President’s inspiring vision of American technological leadership stands in stark contrast to the current realities of the AI industry. Behind the polished demos and ambitious promises of the very companies President Trump is partnering with lies a complex reality: an army of hidden human workers in developing countries keeping these “automated” systems running. The wanting monster that is Silicon Valley’s insatiable appetite for growth has created a global system of outsourcing that challenges the premise of domestic job creation.
The Automation Illusion
“Mr. Banks began to wriggle. His heart began to jiggle. A little note of misery sounded in his mind. What could possibly be wrong? It was a perfect day for as snooze by the stream. But now he wanted something else, something more.” Martine Murray, The Wanting Monster
Like Mr. Banks suddenly dissatisfied with his peaceful afternoon by the stream, corporate America has been seized by a restless hunger for AI transformation that promises everything but delivers far less than advertised. The reality behind AI’s promises becomes stark when examining the latest industry data, which reveals a landscape littered with abandoned projects and unmet expectations. Research from Boston Consulting Group (2024) found that 74% of companies struggle to achieve and scale value from AI initiatives, while between 70% and 85% of current AI initiatives fail to meet their expected outcomes.
The scope of this failure becomes even more troubling when viewed through the lens of rigorous academic research. RAND Corporation’s investigation (Ryseff et al., 2024), which involved interviewing 65 data scientists and engineers with hands-on experience, found that by some estimates, more than 80 percent of AI projects fail. This represents twice the rate of failure seen in information technology projects that do not involve AI, pointing to something inherently problematic about how artificial intelligence is being implemented across industries. The RAND study’s most valuable contribution lies in its identification of five leading root causes that consistently undermine AI initiatives:
- Misunderstandings and miscommunications about the intent and purpose of the project.
- Lack of alignment between the technical staff’s understanding and the project goals.
- Insufficient time commitment – projects require at least a year of dedicated focus.
- Technology-focused approach rather than problem-focused approach.
- Inadequate infrastructure to manage data and deploy completed AI models. RAND report (Ryseff et al., 2024)
Insatiable Hunger and Resource Consumption
“Soon it was only a trickle. The fish gasped and flapped, the frogs jumped away, and the reeds withered and died.” Martine Murray, The Wanting Monster
Like Mr. Banks (The Wanting Monster) gazing at the shimmering stream and suddenly needing to possess it entirely, Silicon Valley’s tech giants have developed an insatiable hunger for computational power that grows with each feeding. The wanting monster’s whisper has led them down a path where Google’s greenhouse gas emissions have surged by 48% since 2019, driven by an ever-expanding appetite for data centers and supply chains that stretch across continents (Hanna & Berry, 2024). The company now acknowledges that integrating AI into products makes reducing emissions challenging and has quietly abandoned its previous carbon neutrality claims, much like Mr. Banks abandoning his peaceful afternoon by the stream (Hanna & Berry, 2024).
Microsoft, too, has felt the monster’s pull, reporting a 30% increase in emissions since 2020 as data center expansion accelerates. Yet the wanting only intensifies. Google plans to spend $75 billion on AI infrastructure alone in 2025, while Apple has announced a staggering $500 billion commitment to manufacturing and data centers over the next four years (International Energy Agency, 2025). The scale of this digital consumption defies comprehension. Global electricity demand for data centers is projected to double by 2030, reaching 945 TWh and representing nearly 3% of all global electricity consumption (International Energy Agency, 2025). A typical AI data center consumes as much power as 100,000 households, but the largest facilities under construction will demand 20 times that amount. Like the villagers who built ever-larger ladders to reach more stars, each new data center must be bigger, more powerful, more resource-hungry than the last.
Beneath this digital wanting lies a web of global dependencies that would make the monster smile. As the International Energy Agency (2025) notes:
(IIEA, 2025)
“The supply chains for the components going into data centers are complex and globalized. For example, gallium is an increasingly critical metal used in cutting-edge computer chips and power electronics, offering significant efficiency benefits compared with traditional silicon-based semiconductor designs. China currently accounts for around 99% of global refined gallium supply. Our estimates indicate that in 2030, demand for gallium for data centres could reach over 10% of today’s supply.”
Even individual acts of digital consumption carry surprising weight. Each ChatGPT query demands 7 – 9 watt hours of energy, consuming 23 to 30 times more power than a simple Google search (De Vries, 2025). The cooling systems alone require the equivalent of a bottle of water for every 10-50 ChatGPT responses (Washington Post, 2024). What appears as effortless automation to users represents a hidden torrent of resource consumption.
Implementation Meets Reality
“Mrs. Walton began to frown and fret. She was irritated. Why was she picking flowers for Maria when it was really she herself who deserved them? She should fill her own house with flowers. Yes, she should have the most fragrant, the most colorful, the most stylish house in the whole village.” Martine Murray, The Wanting Monster
Like Mrs. Walton’s sudden dissatisfaction with her modest flower gathering, corporate AI ambitions have grown far beyond what current technology can actually deliver. The wanting monster whispers promises of automation perfection, but reality tells a different story when these systems encounter the messy complexities of human interaction.
Air Canada discovered this harsh truth when their chatbot confidently advised a grieving customer to follow a bereavement refund policy that simply did not exist, ultimately leading to legal consequences when the customer sued (Olavsrud, 2025). The automated system, designed to handle customer inquiries efficiently, had fabricated policy details with the same confidence it might display when providing accurate information.
Google’s own wanting monster moment arrived in May 2024 when their AI Overviews feature began offering users bizarre suggestions, including the now-infamous recommendation to glue cheese to pizza (Rhiannon, W., MIT Review, 2025). What was intended to be the crown jewel of search innovation instead became a cautionary tale about the gap between technological aspiration and practical readiness.
The legal ramifications can be severe when these systems fail. iTutor Group learned this when they agreed to pay $365,000 to settle a lawsuit after their AI-powered recruiting software systematically rejected qualified candidates based solely on age and gender (Olavsrud, 2025). The federal Equal Employment Opportunity Commission made clear that technological automation does not absolve companies of responsibility, stating that “even when technology automates the discrimination, the employer is still responsible.” The wanting monster’s promise of effortless automation continues to seduce, but implementation reveals that the most beautiful technological flowers still require human hands to tend them properly.
The America First Agenda vs the Global Workforce Model
“More and more ladders rise up and the sky soon grows starless. With the stream gone and the flowers gone and the forest gone, with the birds silent and the bees still, this tranquil little world finds itself unworlded.” Martine Murray, The Wanting Monster
In the village of the wanting monster, the residents climbed higher and higher ladders to pluck stars from the sky, each believing they deserved more than their neighbors. Silicon Valley’s pursuit of AI dominance follows a remarkably similar pattern, though instead of ladders reaching toward stars, a vast network of digital connections reaches toward workers across the globe who make the magic of “automation” possible.
Behind every AI system that appears to think, learn, and respond lies an intricate human infrastructure that remains deliberately invisible. As AI researcher Milagros Miceli in Netzpolitik observes,
“Human labor is a necessary part of the loop to generate and maximize surplus value. But for this, labor needs to be available and cheap. Hence, most tech giants rely on platforms and companies that provide an outsourced workforce, available 24/7 at low costs”
Milagros Miceli in Netzpolitik
Like the wanting monster’s whispered promises that made each villager crave more than they had, the promise of pure automation masks a dependency on human intelligence that operates far from Silicon Valley’s gleaming offices.
The geography of this workforce tells its own story. Data workers in Kenya earn as little as $2 per hour, while their counterparts in Argentina receive $1.7 per hour, often without the employment protections that workers in wealthier nations might expect (Data Workers’ Inquiry, 2024). In the Philippines, workers spend their days labeling data for multi-billion-dollar companies like Scale AI, frequently earning below local minimum wage standards (The Conversation, 2024). Each labeled image, each moderated comment, each classified data point represents a human decision dressed up as machine intelligence.
This global workforce model represents exactly the kind of job displacement that America has struggled with for decades. These AI-supporting roles involving data labeling, content moderation, and algorithm training could provide meaningful employment opportunities for American workers who have seen manufacturing and service jobs move overseas. The scale of this hidden infrastructure is staggering, resembling the village after all the stars had been plucked from the sky. Tech giants like Meta, Google, OpenAI, and Microsoft have built extensive networks of data labeling operations spanning the Philippines, Kenya, India, Pakistan, and Colombia. Scale AI alone, a California company now valued at $14 billion, employs at least 10,000 workers in the Philippines (The Conversation, 2024). Each worker represents a small piece of the human intelligence that powers systems marketed as artificially intelligent.
Like the village that found itself “unworlded” after the wanting monster’s influence had spread, the current AI landscape has created a world where essential technical work has been shipped overseas. The challenge ahead lies in recognizing that AI systems require substantial human intelligence and bringing that work back to where it can support American families and communities. The stars may have been plucked from the sky, but they should shine brightest when powering economic opportunity at home.
The Need for Human-in-the-Loop Systems
“Billie Ray cupped her other hand to make a roof, and then she wandered toward the dry river bed, where she sat on its banks and began to rock her hand and sing the lullaby her mother had once sung to her. No one had ever sung to the Wanting Monster before. Nor had it ever been cared for.” Martine Murray, The Wanting Monster
Just as young Billie Ray recognized that the wanting monster needed care rather than conquest, the author believes that the AI industry should acknowledge a fundamental truth: human oversight remains essential for AI systems to function reliably, and perhaps more importantly, these human roles represent exactly the kind of high-value work that should be returning to American shores.
The gap between automation promises and practical reality demonstrates why human-in-the-loop approaches may be necessary for successful AI implementation. But unlike the current model that scatters this essential work across distant countries, an America First approach to AI recognizes these roles as opportunities to rebuild domestic technical expertise. AI systems require human judgment to moderate content, verify outputs, and guide algorithmic decisions, and such tech positions could become the foundation for a new generation of American technology workers.
American leadership in AI should include transparency requirements by large companies serving the domestic market that reveal the true human workforce behind these systems, helping bridge the gap between AI promises and operational realities while ensuring that the substantial human intelligence required by AI systems benefits American workers and communities where it belongs.
Revitalizing America’s Technology Workforce
“What had been withheld was released; what had dried up, flowed. What had hardened was becoming soft again. People unpack their suitcases, take the stars out of their pockets, and set about collecting seeds, tilling the ground, and filling watering cans to replant the trees and flowers.” Martine Murray, The Wanting Monster
In the fable’s resolution, the villagers stopped chasing impossible dreams and began the patient work of restoration. They returned the stars to the sky, replanted their forests, and tended their streams back to life. Like the wanting monster in the fable, Silicon Valley’s appetite for growth and dominance has created challenges that require careful attention. The industry’s current model relies on global networks and human labor in ways that may not align with promises of full automation, domestic job creation, or national security interests. But just as the villagers learned to work with their environment rather than against it, American AI development can learn to work with human intelligence rather than trying to eliminate it.
President Trump’s Stargate vision represents a chance to reverse decades of technology job displacement. Instead of building data centers that depend on distant workforces, America could create AI infrastructure supported by domestic talent. The technical roles that keep AI systems functioning, the human judgment that ensures they operate safely, and the specialized knowledge required to improve them could all become part of a revitalized American technology workforce.
The children in the story wept for their world, but they also helped restore it. Today’s challenge is ensuring that technological progress serves American workers and communities, creating systems that harness AI while strengthening human opportunity. The author believes that the stars shine brightest when they illuminate pathways home.
References and further reading.
Axios. (2025, May 29). Trump’s AI embrace risks crossing MAGA’s vulnerable working class. Axios. Retrieved from https://www.axios.com/2025/05/29/ai-trump-maga-job-losses
Boston Consulting Group. (2024, October). AI adoption in 2024: 74% of companies struggle to achieve and scale value. BCG Press Release. Retrieved from https://www.bcg.com/press/24october2024-ai-adoption-in-2024-74-of-companies-struggle-to-achieve-and-scale-value
Brookings Institution. (2024, November 26). Moving toward truly responsible AI development in the global AI market. Brookings. Retrieved from https://www.brookings.edu/articles/moving-toward-truly-responsible-ai-development-in-the-global-ai-market/
Content moderation using AI and human moderators. (2024, May 21). Elv.ai. Retrieved from https://elv.ai/
Data Workers’ Inquiry. (2024, July 8). The hidden workers behind AI tell their stories. Netzpolitik. Retrieved from https://netzpolitik.org/2024/data-workers-inquiry-the-hidden-workers-behind-ai-tell-their-stories/
De Vries, A. (2025). How much energy will AI really consume? The good, the bad and the unknown. Nature. Retrieved from https://www.nature.com/articles/d41586-025-00616-z
Environmental Impact of Google-Microsoft’s battle for AI. (2023, September 27). Karmametrix. Retrieved from https://karmametrix.com/blog/web-sustainability/google-vs-microsoft-the-battle-for-ai-environmental-impact/
Hanna, A., & Berry, F. (2024, July 12). AI brings soaring emissions for Google and Microsoft, a major contributor to climate change. NPR. Retrieved from https://www.npr.org/2024/07/12/g-s1-9545/ai-brings-soaring-emissions-for-google-and-microsoft-a-major-contributor-to-climate-change
IEA (2025), Energy and AI, IEA, Paris https://www.iea.org/reports/energy-and-ai, Licence: CC BY 4.0. Retrieved from https://www.iea.org/reports/energy-and-ai/executive-summary
Low-paid workers in poor countries label data for AI giants. (2023, October 16). DailyAlts. Retrieved from https://dailyalts.com/low-paid-workers-in-poor-countries-label-data-for-ai-tech-giants/
Massachusetts Institute of Technology. (2025, January 17). Explained: Generative AI’s environmental impact. MIT News. Retrieved from https://news.mit.edu/2025/explained-generative-ai-environmental-impact-0117
Murray, M., & Read, A. L. (2021). The wanting monster. Enchanted Lion Books.
OpenAI operating costs. (2024). The financial challenges of leading in AI: A look at OpenAI’s operating costs. Renewable AI. Retrieved from https://renewableai.org/news/the-financial-challenges-of-leading-in-ai-a-look-at-openais-operating-costs/
Olavsrud, T. (2025, March 21). 12 famous AI disasters. CIO. Retrieved from https://www.cio.com/article/190888/5-famous-analytics-and-ai-disasters.html
Reuters. (2024, October 11). ByteDance’s TikTok cuts hundreds of jobs in shift towards AI content moderation. Reuters. Retrieved from https://www.reuters.com/technology/bytedance-cuts-over-700-jobs-malaysia-shift-towards-ai-moderation-sources-say-2024-10-11/
Reuters. (2025, January 22). Trump announces private-sector $500 billion investment in AI infrastructure. Reuters. Retrieved from https://www.reuters.com/technology/artificial-intelligence/trump-announce-private-sector-ai-infrastructure-investment-cbs-reports-2025-01-21/
Ryseff, J., De Bruhl, B. F., & Newberry, S. J. (2024). The root causes of failure for artificial intelligence projects and how they can succeed: Avoiding the anti-patterns of AI. RAND Corporation. Retrieved from https://www.rand.org/pubs/research_reports/RRA2680-1.html
Smith, A., Chandran, R., & Ramos, M. (2023, March 20). AI boom is dream and nightmare for workers in developing countries. The Japan Times. Retrieved from https://www.japantimes.co.jp/news/2023/03/20/business/tech/ai-boom-dream-nightmare/
The Conversation. (2024, November 21). AI is a multi-billion dollar industry. It’s underpinned by an invisible and exploited workforce. The Conversation. Retrieved from https://theconversation.com/ai-is-a-multi-billion-dollar-industry-its-underpinned-by-an-invisible-and-exploited-workforce-240568
Rhiannon, W., (2025, January 10). The biggest AI flops of 2024. MIT Technology Review. Retrieved from https://www.technologyreview.com/2024/12/31/1109612/biggest-worst-ai-artificial-intelligence-flops-fails-2024/
Time. (2025, April 9). Trump wants tariffs to bring back U.S. jobs. They might speed up AI automation instead. Time. Retrieved from https://time.com/7276087/trump-tariffs-ai-automation-robots/
University of Technology Sydney. (2023, November 17). Analysis: Long hours and low wages: The human labour powering AI’s development. Brighter World. Retrieved from https://brighterworld.mcmaster.ca/articles/analysis-long-hours-and-low-wages-the-human-labour-powering-ais-development/
Washington Post. (2024, October 11). A bottle of water per email: The hidden environmental costs of using AI chatbots. The Washington Post. Retrieved from https://www.washingtonpost.com/technology/2024/09/18/energy-ai-use-electricity-water-data-centers/
White House. (2025, January 24). Fact sheet: President Donald J. Trump takes action to enhance America’s AI leadership. The White House. Retrieved from https://www.whitehouse.gov/fact-sheets/2025/01/fact-sheet-president-donald-j-trump-takes-action-to-enhance-americas-ai-leadership
Dr. Jasmin (Bey) Cowin, a columnist for Stankevicius, employs the ethical framework of Nicomachean Ethics to examine how AI and emerging technologies shape human potential. Her analysis explores the risks and opportunities that arise from tech trends, offering personal perspectives on the interplay between innovation and ethical values. Connect with her on LinkedIn.